The fundamental question that people have about early retirement is: how much do I need to save to retire?
For those who understand the mathematics behind early retirement, the main thing that determines how big our retirement portfolio must be is the safe withdrawal rate.
Essentially, safe withdrawal rate is the amount of money you can withdraw from your portfolio when you are retired without running out of money.
I stumbled upon an amazing 28 part series on Early Retirement Now that calculates safe withdrawal rates for U.S. markets. It calculates safe withdrawal rates for different retirement lengths, asset allocation, and more!
However, there are major differences between U.S. and Australian markets that mean each would likely have different safe withdrawal rates.
It’s not fair that our U.S. friends have such great data on their required retirement portfolio and we don’t. So in the next few posts, I plan to replicate some of the analysis on Early Retirement Now — with Australian data, for Australians.
Early Retirement Now’s Safe Withdrawal Rate Guide
The most comprehensive and well-explained analysis of early retirement was written by Kartsen at Early Retirement Now.
In his mammoth 28 part safe withdrawal rate series, Karsten from Early Retirement Now breaks down the mathematics behind the 4% rule. He found some very interesting things:
- The 4% rule should really be called the 4% rule of thumb. There is no one-size-fits all approach. Not only is every retirement case unique (e.g. how long you’ll be retired for), but withdrawal rates also depend on the equity valuations and bond yields at the time of retirement and beyond.
- Bonds are less appropriate as the retirement length grows. Although the Trinity Study favours a high bond allocation, Karsten found that bonds decrease success rate for longer retirement lengths. This sounds counterintuitive but if you have bad luck with low equity returns at the beginning of retirement, you need a higher proportion of equities to ‘make up’ those lost returns (bond returns are too low).
- Sequence of return risk is scary! The biggest cause of retirement portfolios failing is a long-ish period of negative returns in the beginning of retirement that significantly deplete capital. Without significantly higher returns later on, the portfolio will be too depleted to last 30 or more years. That being said, we can also benefit from the reverse case (i.e. a long-ish period of high returns at the beginning of retirement).
- Government benefits don’t help. Karsten analysed the U.S. Social Security regime and found that government benefits do not make a significant difference in safe withdrawal rate. Given that the Australian government pension is means-tested, I’d guess that would be the case of Australians too.
That leads me to the next point: All of Karsten’s analysis was done using U.S. equity and bond data, and thus applies to Americans.
But what would a safe withdrawal rate be for Australians?
Over the next few months, we will replicate a number of analyses that Karsten conduct on Early Retirement Now. But this time we’ll use Australian data.
What analysis will we run?
The plan is to start with five posts that focus on different components of safe withdrawal rates. These posts will reflect some of the major findings from Early Retirement Now.
I’m sure that answering the following questions will lead to more questions. So there’s plenty of opportunity for looking at more things are our series progresses.
1. What is a safe withdrawal rate for Australians?
Most of us are aware of the Trinity Study. The Trinity study looked at how much a retiree could withdraw from their retirement portfolio per year without exhausting all of the money in their portfolio.
Basically, the study found that a retiree with a portfolio that had a 50/50 mix of stocks and bonds could withdraw 4% of their initial portfolio balance per year (adjusted for inflation) and still have some money left after 30 years.
In his Ultimate Guide to Safe Withdrawal Rates: Part 2 article, Karsten replicates the 4% rule with his own data, so we can be pretty confident in that data point.
But what about Australia?
We already know that safe withdrawal rates are determined by risk and return. And we also know that the risk and return characteristics of Australian equities and bonds are different than the U.S. Therefore, it follows that Australian safe withdrawal rates should be different than U.S. safe withdrawal rates
2. How does length of retirement affect safe withdrawal rates?
The original Trinity Study looked at retirement lengths of 15, 20, 25 and 30 years. Assuming that we die around 85 years old, these retirement lengths are only really relevant for people who are retiring at age 55 and beyond.
In the early retirement community, there are plenty of people who have retired in their early thirties. These people will be retired for up to 50 or 60 years!
Research has already proven that the safe withdrawal rate decreases for longer retirement periods. And this is a scary thought for people who will be retired half their life.
The Early Retirement Now analysis showed that if you invested your funds in U.S. equities and bonds markets, a withdrawal rate of 3.5% and at least 75% equities maintains good portfolio success rates over 60 years.
It goes without saying that longer retirement lengths require lower withdrawal rates in Australia too. But we don’t know if Australian safe withdrawal rates are more sensitive or less sensitive to retirement length than U.S. retirement portfolios.
3. Asset allocation and safe withdrawal rates for Australians
The quantitative difference between equities and bonds is that equities are higher risk (i.e. higher volatility) and higher reward (i.e. higher average returns). By choosing to invest in equities you are accepting a bumpy ride with big up/down swings. But if you can stomach those swings, you should get a higher overall return.
The Trinity study showed that for shorter retirement lengths (e.g. 20-30 years) you can get away with up to 50% of your retirement portfolio invested in bonds. Early Retirement Now subsequently showed that longer retirement lengths require a higher equity investment.
Both of these studies looked at various U.S. equity and U.S. bond allocations. This is probably because many Americans invest exclusively in U.S. markets.
However, it’s common for Australians to invest in a mix of domestics markets and international markets.
We plan to extend the Trinity and Early Retirement Now studies by looking asset allocations comprising a mix of Australian equities, Australian bonds, U.S. equities and U.S. bonds. Each of these markets have different risk/return characteristics and should affect safe withdrawal rates in different ways.
4. Should Australians be afraid of sequence of returns risk?
Sequence of return risk has been called the most significant threat to safe withdrawal rates since WWII. Ok, that’s a lie — but it’s a very significant risk!
The Early Retirement Now study also unpacked the implications of sequence of returns. Karsten found that sequence of return was a bigger influence on safe withdrawal rates than average portfolio returns.
Here’s an example of how sequence or return risk can bite: you accumulate your portfolio during a bull market and retire right as the market drops 50%. Your portfolio balance immediately halves just before you need to use it – ouch!
According to the mathematics behind sequence of return risk, the risk is higher for markets with higher volatility. Therefore, it’s higher for equities than for bonds.
5. The Australian taxation system vs safe withdrawal rates
One of the criticisms of most early retirement calculations is that they don’t take into account the effect of tax.
For example, if you plan to live off $75,000 by selling down part of your portfolio after you retire, you’ll need to pay tax on that income. And to take home $75,000 after tax might require withdrawing $85,000 or more out of your portfolio.
Of course, calculating the exactly tax implications of any given portfolio is difficult. Everybody’s circumstances are different. So let’s look at the big structural contributor to retirement and taxation: superannuation.
Now we need to collate a database of Australian and U.S. equities and bonds, and then validate it for accuracy.
Although it seems reasonably straightforward, it’s extremely important that we use accurate data — so I’ll spend a bit of time making sure the data is high quality.
In the next post, I’ll share how I created the dataset and how we validated the accuracy by comparing it against similar datasets.
Until then, all the best!